Marie V. Brasseur’s research while affiliated with University of Duisburg-Essen and other places

Urbanization and the resulting modifications of freshwater ecosystems can play an important role in shaping metapopulation structure and dynamics of aquatic organisms. Ecological restoration aims at improving river ecosystems by reducing or removing anthropogenic stressors and habitat fragmentation, facilitating natural dispersal among population patches. However, the success of such ecological restoration measures is not guaranteed, and for many of the functionally important but smaller organisms, improved connectivity is difficult to assess. Here, genetic markers can help in assessing small‐scale connectivity and in identifying persisting gene flow barriers. In this study used high‐resolution genetic markers to study the metapopulation structure of two ecologically important amphipod species, Gammarus pulex and Gammarus fossarum, in the heavily urbanized Emscher catchment in Germany. This catchment was strongly degraded and polluted for over a century but has been restored over the past two decades. For both strictly aquatic species, we analyzed mitochondrial cytochrome c oxidase I (COI) gene sequences as well as nuclear genome‐wide single nucleotide polymorphism (SNP) data. We detected strong metapopulation structure within both species, which was mainly driven by catchment affiliation, wastewater, large in‐stream barriers, and recent recolonization of restored stream sections. However, population structure was not fully explained by these factors, indicating that eco‐evolutionary factors such as priority effects, adaptation, or biotic interactions play a role in shaping the population structure. Furthermore, our data show a strong mito‐nuclear discordance for both species with regard to detailed population structure and also the presence of possible cryptic species for G. pulex. Here, nuclear data indicate that the diverging mitochondrial lineages of G. pulex (Gp‐C and Gp‐E) represent only one species in this region. Our study shows how genetic markers can support the assessment of population connectivity and thus evaluate the success of ecological restoration.

The viviparous eelpout Zoarces viviparus is a common fish across the North-East Atlantic and has successfully colonized habitats across environmental gradients. Due to its wide distribution and predictable phenotypic responses to pollution, Z. viviparus is used as bioindicator organism and has been routinely sampled over decades by several countries to monitor marine environmental health. Additionally, this species is a promising model to study adaptive processes related to environmental change, specifically global warming. Here, we report the chromosome-level genome assembly of Z. viviparus, which has a size of 663 mega base pairs (mbp) and consists of 607 scaffolds (N50 = 26 mbp). The 24 largest represent the 24 chromosomes of the haploid Z. viviparus genome, which harbors 98% of the complete BUSCOs defined for ray-finned fish, indicating that the assembly is highly contiguous and complete. Comparative analyses between the Z. viviparus assembly and chromosome-level genomes of two other eelpout species revealed a high synteny, but also an accumulation of repetitive elements in the Z. viviparus genome. Our reference genome will be an important resource enabling future in-depth genomic analyses of the effects of environmental change in this important bioindicator species.

Intensive agriculture drives an ongoing deterioration of stream biodiversity and ecosystem functioning across the planet. Key agricultural stressors include increased deposited fine sediment and insecticides flushed from adjacent land into streams. The individual and combined effects on aquatic biota are increasingly studied, but the functional consequences of biodiversity loss associated with agricultural stressors remain poorly understood. We addressed this knowledge gap by examining the effects of fine sediment and different concentrations of the insecticide chlorantraniliprole on organic‐matter decomposition. We conducted an outdoor stream mesocosm experiment. Mesocosms contained a standardised organic‐matter assay (the cotton‐strip assay), which was used to assess organic‐matter decomposition rates (as tensile‐strength loss of the fabric) and microbial respiration of the cotton strips. The decomposition rate of strips buried under fine sediment was inhibited, a result we attribute to the limited accessibility for invertebrate feeding and microbial activities, as well as the limited nutrient and dissolved oxygen exchange. The insecticide also inhibited decay rates, a finding we relate to reduced invertebrate grazing and associated excessive algal growth. In contrast to decomposition rates, we did not observe stressor effects on microbial respiration. An interaction between fine sediment and chlorantraniliprole was not identified. Our results suggest that stressors induced by agriculture affect functions of stream ecosystems through a variety of pathways and operate by modifying habitats and biotic interactions. By examining a combination of stressors and responses that have not been addressed before, this study gives important insights into the effects of agricultural practices on streams. Understanding the effects of chlorantraniliprole is especially important since it is likely to become more widely used in future agricultural practice due to the increasing ban on neonicotinoid insecticides. Furthermore, most experimental studies address multiple stressor effects on biota. For a comprehensive understanding of complex stressor effects on ecosystems, ecosystem functions also need to be studied, such as the organic‐matter decomposition within streams.

Introduction Microbes are increasingly (re)considered for environmental assessments because they are powerful indicators for the health of ecosystems. The complexity of microbial communities necessitates powerful novel tools to derive conclusions for environmental decision-makers, and machine learning is a promising option in that context. While amplicon sequencing is typically applied to assess microbial communities, metagenomics and total RNA sequencing (herein summarized as omics-based methods) can provide a more holistic picture of microbial biodiversity at sufficient sequencing depths. Despite this advantage, amplicon sequencing and omics-based methods have not yet been compared for taxonomy-based environmental assessments with machine learning. Methods In this study, we applied 16S and ITS-2 sequencing, metagenomics, and total RNA sequencing to samples from a stream mesocosm experiment that investigated the impacts of two aquatic stressors, insecticide and increased fine sediment deposition, on stream biodiversity. We processed the data using similarity clustering and denoising (only applicable to amplicon sequencing) as well as multiple taxonomic levels, data types, feature selection, and machine learning algorithms and evaluated the stressor prediction performance of each generated model for a total of 1,536 evaluated combinations of taxonomic datasets and data-processing methods. Results Sequencing and data-processing methods had a substantial impact on stressor prediction. While omics-based methods detected a higher diversity of taxa than amplicon sequencing, 16S sequencing outperformed all other sequencing methods in terms of stressor prediction based on the Matthews Correlation Coefficient. However, even the highest observed performance for 16S sequencing was still only moderate. Omics-based methods performed poorly overall, but this was likely due to insufficient sequencing depth. Data types had no impact on performance while feature selection significantly improved performance for omics-based methods but not for amplicon sequencing. Discussion We conclude that amplicon sequencing might be a better candidate for machine-learning-based environmental stressor prediction than omics-based methods, but the latter require further research at higher sequencing depths to confirm this conclusion. More sampling could improve stressor prediction performance, and while this was not possible in the context of our study, thousands of sampling sites are monitored for routine environmental assessments, providing an ideal framework to further refine the approach for possible implementation in environmental diagnostics.

Background Freshwater ecosystem degradation and biodiversity decline are strongly associated with intensive agricultural practices. Simultaneously occurring agricultural stressors can interact in complex ways, preventing an accurate prediction of their combined effects on aquatic biota. Here, we address the limited mechanistic understanding of multiple stressor effects of two globally important stressors, an insecticide (chlorantraniliprole), and increased fine sediment load and assessed their impact on the transcriptomic profile of two stream macroinvertebrates: the amphipod Gammarus pulex and the caddisfly Lepidostoma basale. Results We identified mainly antagonistic stressor interactions at the transcriptional level, presumably because the insecticide adsorbed to fine sediment particles. L. basale, which is phylogenetically more closely related to the insecticide’s target taxon Lepidoptera, exhibited strong transcriptional changes when the insecticide stressor was applied, whereas no clear response patterns were observed in the amphipod G. pulex. These differences in species vulnerability can presumably be attributed to molecular mechanisms determining the cellular affinity toward a stressor as well as differential exposure patterns resulting from varying ecological requirements between L. basale and G. pulex. Interestingly, the transcriptional response induced by insecticide exposure in L. basale was not associated with a disruption of the calcium homeostasis, which is the described mode of action for chlorantraniliprole. Instead, immune responses and alterations of the developmental program appear to play a more significant role. Conclusions Our study shows how transcriptomic data can be used to identify multiple stressor effects and to explore the molecular mechanisms underlying stressor-induced physiological responses. As such, stressor effects assessed at the molecular level can inform about modes of action of chemicals and their interplay with non-chemical stressors. We demonstrated that stressor effects vary between different organismic groups and that insecticide effects are not necessarily covered by their described mode of action, which has important implications for environmental risk assessment of insecticides in non-target organisms.

Pesticides are major agricultural stressors for freshwater species. Exposure to pesticides can disrupt the biotic integrity of freshwater ecosystems and impair associated ecosystem functions. Unfortunately, physiological mechanisms through which pesticides affect aquatic organisms are largely unknown. For example, the widely-used insecticide chlorantraniliprole is supposed to be highly selective for target pest species, i.e. Lepidoptera (butterflies), but its effect in aquatic non-target taxa is poorly studied. Using RNA-sequencing data, we quantified the insecticide effect on three aquatic invertebrate species: the caddisfly Lepidostoma basale, the mayfly Ephemera danica and the amphipod Gammarus pulex. Further, we tested how the insecticide-induced transcriptional response is modulated by biotic interaction between the two leaf-shredding species L. basale and G. pulex. While G. pulex was only weakly affected by chlorantraniliprole exposure, we detected strong transcriptional responses in L. basale and E. danica, implying that the stressor receptors are conserved between the target taxon Lepidoptera and other insect groups. We found in both insect species evidence for alterations of the developmental program. If transcriptional changes in the developmental program induce alterations in emergence phenology, pronounced effects on food web dynamics in a cross-ecosystem context are expected.

Reliable biodiversity data are crucial for environmental research and management. Unfortunately, data paucity prevails for many regions and organismal groups such as aquatic invertebrates. High-throughput DNA-based identification, in particular DNA metabarcoding, has accelerated biodiversity data generation. However, in the process of metabarcoding, specimens are usually destroyed, precluding later specimen-based analyses. Metabarcoding of DNA released into the preservative ethanol has been proposed as a non-destructive alternative, but proof-of-principle studies have yielded ambiguous results, reporting variance in detection probability for various taxa and methodological biases. In this study, we tested the performance of preservative-based metabarcoding of aquatic invertebrates in comparison to a standard morpho-taxonomic assessment based on samples from one of Europe’s last free-flowing rivers, the Vjosa. Multi-habitat samples were collected at 43 sites in two seasons and stored in ethanol, after fixation in formaldehyde for morpho-taxonomic analyses. Preservative-based DNA metabarcoding detected three times more taxa but failed to detect other taxa found using the standard method. In addition to incomplete reference data and primer bias that likely precluded the detection of specific taxa, preservative-based DNA metabarcoding cannot provide accurate abundance estimates. However, the metabarcoding data revealed distinct small-scale and large-scale community patterns in the Vjosa river network, which were also recovered by quantitative data of the standard approach. Overall, our results indicate that preservative-based metabarcoding provides important biodiversity data, which could be further improved by quantitative validation. The method is robust and reliable, even though samples were taken under harsh field-conditions and stored without cooling. Further, our results emphasise the need for reliable DNA barcoding reference libraries. Building those may be supported by preservative-based metabarcoding that maintains intact vouchers for subsequent specimen-based analyses.

Mitochondrial DNA (mtDNA) sequences are often found as byproducts in next‐generation sequencing (NGS) datasets that were originally created to capture genomic or transcriptomic information of an organism. These mtDNA sequences are often discarded, wasting this valuable sequencing information. We developed MitoGeneExtractor, an innovative tool which allows to extract mitochondrial protein coding genes (PCGs) of interest from NGS libraries through multiple sequence alignments of sequencing reads to amino acid references. General references, for example on order level are sufficient for mining mitochondrial PCGs. In a case study, we applied MitoGeneExtractor to recently published genomic datasets of 1993 birds and were able to extract complete or nearly complete sequences for all 13 mitochondrial PCGs for a large proportion of libraries. Compared to an existing assembly guided sequence reconstruction algorithm, MitoGeneExtractor was faster and substantially more sensitive. We compared COI sequences mined with MitoGeneExtractor to COI databases. Mined sequences show a high sequence similarity and correct taxonomic assignment between the recovered sequence and the assigned morphospecies in most samples. In some cases of incongruent taxonomic assignments, we found evidence for contamination in NGS libraries. MitoGeneExtractor allows a fast extraction of mitochondrial PCGs from a wide range of NGS datasets. We recommend to routinely harvest and curate mitochondrial sequence information from genomic resources. MitoGeneExtractor output can be used to identify contaminated NGS libraries and to validate the species identity of the sequenced animal based on the extracted COI sequences.

Background Freshwaters are exposed to multiple anthropogenic stressors, leading to habitat degradation and biodiversity decline. In particular, agricultural stressors are known to result in decreased abundances and community shifts towards more tolerant taxa. However, the combined effects of stressors are difficult to predict as they can interact in complex ways, leading to enhanced (synergistic) or decreased (antagonistic) response patterns. Furthermore, stress responses may remain undetected if only the abundance changes in ecological experiments are considered, as organisms may have physiological protective pathways to counteract stressor effects. Therefore, we here used transcriptome-wide sequencing data to quantify single and combined effects of elevated fine sediment deposition, increased salinity and reduced flow velocity on the gene expression of the amphipod Gammarus fossarum in a mesocosm field experiment. Results Stressor exposure resulted in a strong transcriptional suppression of genes involved in metabolic and energy consuming cellular processes, indicating that G . fossarum responds to stressor exposure by directing energy to vitally essential processes. Treatments involving increased salinity induced by far the strongest transcriptional response, contrasting the observed abundance patterns where no effect was detected. Specifically, increased salinity induced the expression of detoxification enzymes and ion transporter genes, which control the membrane permeability of sodium, potassium or chloride. Stressor interactions at the physiological level were mainly antagonistic, such as the combined effect of increased fine sediment and reduced flow velocity. The compensation of the fine sediment induced effect by reduced flow velocity is in line with observations based on specimen abundance data. Conclusions Our findings show that gene expression data provide new mechanistic insights in responses of freshwater organisms to multiple anthropogenic stressors. The assessment of stressor effects at the transcriptomic level and its integration with stressor effects at the level of specimen abundances significantly contribute to our understanding of multiple stressor effects in freshwater ecosystems.

Background: Microbes are increasingly (re)considered for environmental assessments because they are powerful indicators for the health of ecosystems. The complexity of microbial communities necessitates powerful novel tools to derive conclusions for environmental decision-makers, and machine learning is a promising option in that context. While amplicon sequencing is typically applied to assess microbial communities, metagenomics and total RNA sequencing (herein summarized as omics-based methods) can provide a more holistic picture of microbial biodiversity at sufficient sequencing depths. Despite this advantage, amplicon sequencing and omics-based methods have not yet been compared for taxonomy-based environmental assessments with machine learning. In this study, we applied 16S and ITS-2 sequencing, metagenomics, and total RNA sequencing to samples from a stream mesocosm experiment that investigated the impacts of two aquatic stressors, insecticide and increased fine sediment deposition, on stream biodiversity. We processed the data using similarity clustering and denoising (only applicable to amplicon sequencing) as well as multiple taxonomic levels, data types, feature selection, and machine learning algorithms and evaluated the stressor prediction performance of each generated model for a total of 1,536 evaluated combinations of taxonomic datasets and data-processing methods. Results: Sequencing and data-processing methods had a substantial impact on stressor prediction. While omics-based methods detected much more taxa than amplicon sequencing, 16S sequencing outperformed all other sequencing methods in terms of stressor prediction based on the Matthews Correlation Coefficient. However, even the highest observed performance for 16S sequencing was still only moderate. Omics-based methods performed poorly overall, but this was likely due to insufficient sequencing depth. Data types had no impact on performance while feature selection significantly improved performance for omics-based methods but not for amplicon sequencing. Conclusion: Amplicon sequencing might be a better candidate for machine-learning-based environmental stressor prediction than omics-based methods, but the latter require further research at higher sequencing depths to confirm this conclusion. More sampling could improve stressor prediction performance, and while this was not possible in the context of our study, thousands of sampling sites are monitored for routine environmental assessments, providing an ideal framework to further refine the approach for possible implementation in environmental diagnostics.